engleski

Outcome of the biliary acute pancreatitis is not associated with body mass index(BMI)

We read with interest the recent article by De Waele et al. about impact of body overweight and class I, II and III obesity on the outcome of biliary acute pancreatitis (AP), published in May 2006. issue of Your prestigeous Journal (1). Recently we published in abstract form our results on impact of body weight on the outcome of alcoholic pancreatitis (2). Because we have not found association of severe acute alcoholic pancreatitis with body overweight, we undertook a retrospective survey to analyse our patients with biliary AP and compare results with those published in De Waele study. We retrospectively analysed group of 357 patients with acute biliary pancreatitis admitted to our hospital. The inclusion criteria consisted of combination of clinical features, a typical case history, triple elevation of serum amylase and lipase level from upper limit of normal, and in all patients confirmation of diagnosis by imaging studies which included ultrasound (US) and computed tomography (CT) in all cases, and additionally magnetic resonance cholangiopancreatography (MRCP) or endoscopic ultrasound (EUS) in some cases. The severity of each case was evaluated according to Atlanta criteria (3). We also divided patients in four categories of BMI (Table 1) according to WHO guidelines (4). Statistical analysis was performed using STATISTICA (data analysis software system), StatSoft, Inc. (2005), version 7.1. (www.statsoft.com). The comparison of quantitative data was done using t-test (in the cases of two independent group&#8217; s comparison) or ANOVA (in the case of multiple groups comparison). The frequencies were analyzed by Yates corrected &#61539; 2- test. For determination of the BMI and the AP outcome association, odds ratio (OR) and its 95% confidence interval (CI) were calculated. Logistic regression analysis was employed to confirm or reject the possible association of BMI with the AP outcome. The level of statistical significance was set at 0.05 in all analyses. The two groups of patients with different biliary AP outcome (287 cases with mild and 70 cases with severe AP) didn&#8217; t differ respecting to age (60.1 &#177; ; ; ; ; ; 15.0 years in mild vs 60.1 &#177; ; ; ; ; ; 15.0 in severe AP outcome, p=0.546) and gender distribution (185/102 in mild vs 43/27 female to male ratio in severe AP outcome, p=0.636). Also, the average BMI value was not significantly different in the two groups (28.0 &#177; ; ; ; ; ; 4.8 in mild vs 27.7 &#177; ; ; ; ; ; 4.8 in severe AP, p=0.726). The comparison of the reference group (BMI<25, N=92) and the overweight and obese group (BMI &#61619; 25, N=265) considering age (60.0 &#177; ; ; ; ; ; 18.5 vs 60.5 &#177; ; ; ; ; ; 13.6, p=0.796), female to male gender distribution (59/33 vs 169/96, p=0.948) and mild to severe AP outcome (73/19 vs 214/51, p=0.636), showed no statistical differnece. The OR for the severe outcome in normal and obese patients was calculated as OR=1.09 with 95% CI of 0.51-1.65, therefore being statistically non-significant. Moreover, the groups showed no statistically significant difference with respect to the occurance of complications (local and/or systemic) ; local complications were found in 12 of 92 cases with normal BMI and 51 of 265 cases in overweight and obese group (p=0.520), whereas systemic complications occurred in 12 of 92 cases of normal BMI and in 27 out of 265 cases of overweight and obese patients (p=0.574). Again, OR for the development of local complications in normal and obese patients was 1.33 with 95% CI of 0.67-2.64. Death was noted in the 2 cases out of 92 in normal BMI group and in 4 out of 265 cases of overweight and obese (p=0.965). The overweight and obese group was further subdivided into overweight (25 &#8804; BMI<30), class I obese (30 &#8804; BMI<35) and class II and III obese (BMI&#61619; 35). The complete data and results of statistical comparisons between the four categories of BMI are presented in Table 1. In the overweight group women were present with the frequency lower than expected, while in the class II and III obesety with frequency higher than expected (p=0.004). The biliary AP outcome defined as mild/severe ratio was nonsignificantly different also in this subdivision (p=0.627) and so it was in complications occurance analysis. Odds ratios for the severe outcome in overweight to reference group were 0.85 (CI:0.45-1.61), in class I obese patients to reference group 0.88 (CI:0.41-1.90) and in class II to reference group 1.58 (CI:0.57-4.36), showing no significant association of BMI with biliary AP outcome. Logistic regression analysis of the BMI as a predictor of the severe outcome in biliary AP showed that BMI has no practical predictive value for the AP outcome (M-L testing, &#61539; 2=0.125, p=0.724) in our sample (Figure 1). There are few studies published considering overweight and obesity as a potential risk factor for the development of complications in acute pancreatitis. Conclusions regarding the type of complications according to the Atlanta criteria are not unique among different authors, as well as regarding the mortality rate (5, 6, 7). There&#8217; s only one meta analysis that revealed obesity as a prognostic factor in acute pancreatitis, favouring the development of systemic and local complications, but having no impact on mortality rate (8). Contrary to results of De Waele et al. our study didn&#8217; t show any impact of BMI on the severity and the outcome of patients with acute biliary pancreatitis. Because in our study we analysed even bigger group of obese patients (99 patients with BMI>30), we think that impact of BMI shoud be reconsidered in further studies.

acute biliary pancreatitis; BMI; outcome

nije evidentirano